Towards a hybrid optimization model for elemental content analysis in EDXRF

Author

Ren, Jun ; Liu, Mingzhe ; Tuo, Xianguo ; Li, Zhe ; Shi, Rui

Author_Institution

Coll. of Nucl. Technol. & Autom. Eng.; Chengdu Univ. of Technol., Chengdu Univ. of Technol., Chengdu, China

fYear

2012

fDate

29-31 May 2012

Firstpage

1251

Lastpage

1254

Abstract

This paper presents a hybrid optimization model for predicting the elemental contents such as Ti, V and Fe in energy dispersive X-ray fluorescence (EDXRF) based on least square support vector machine (LS-SVM) and particle swarm optimization (PSO) methods. The model used PSO to optimize LS-SVM parameters. In order to assess the capability and effectiveness of the proposed model, several measurement methods such as SVM model and BP neural network model were compared. The results indicate that the proposed model is feasible for quantitative analysis of elemental contents in nondestructive nuclear measurement applications.

Keywords

X-ray chemical analysis; least squares approximations; nuclear engineering computing; particle swarm optimisation; support vector machines; EDXRF; LS-SVM; PSO methods; elemental content analysis; energy dispersive X-ray fluorescence; hybrid optimization; least square support vector machine; nondestructive nuclear measurement applications; particle swarm optimization; Analytical models; Computational modeling; Iron; Optimization; Particle swarm optimization; Support vector machines; Training; EDXRF; Optimization; Particle swarm optimization; Support vector machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation (ICNC), 2012 Eighth International Conference on

Conference_Location

Chongqing

ISSN

2157-9555

Print_ISBN

978-1-4577-2130-4

Type

conf

DOI

10.1109/ICNC.2012.6234633

Filename

6234633